Hi everyone!
I'm currently investigating transition states and I would like to calculate a 2D or 3D potential energy surface (PES) around the TS structure in order to locate possible saddle points and verify the most probable TS geometry. Does anyone have suggestions on how to set up this type of PES calculation in Gaussian 16?
Many thanks in advance!
Greetings Leonardo Claudio !
The strategy for exploring potential energy surfaces (PES) depends entirely on the specific molecule and transition state being studied, as the PES is fundamentally high-dimensional. In Gaussian, the most practical approach involves performing systematic scans along selected geometric parameters (bond lengths, angles, or torsion angles) while monitoring the corresponding energy changes.
When we refer to dimensionality in PES analysis, we're describing the number of geometric parameters being varied, not the final plot dimensions. A one-dimensional scan, where we vary a single geometric parameter (like a bond length) and track the energy, produces what we commonly call a 1D PES, though this is represented graphically as a 2D plot with energy on the vertical axis and the geometric parameter on the horizontal axis. For more complex analyses, a two-dimensional scan involves varying two geometric parameters simultaneously. In practice, this means first selecting a range for one parameter (say, a bond length), then at each point in that range scanning a second parameter (perhaps a bond angle). The resulting data forms what we term a 2D PES, which we visualize as either a 3D surface plot or a 2D contour plot, with energy represented through color gradients or elevation.
It's crucial to recognize that while the complete PES for any molecule exists in a space of normally 3N-6 dimensions (where N is the number of atoms), comprehensive mapping becomes computationally intractable for all but the simplest systems. Therefore, effective PES exploration requires careful selection of the most chemically relevant degrees of freedom - typically those involved in the reaction coordinate or suspected transition state geometry.
Gaussian provides no automated solution for complete PES generation; instead, you must design your scanning procedures based on chemical intuition and system-specific knowledge. The Gaussian documentation on scanning (https://gaussian.com/scan/) might help you.
Hope my two cents helps.
- Badges
- Science method